Figure 2 | Scientific Reports

Figure 2

From: An automated cell line authentication method for AstraZeneca global cell bank using deep neural networks on brightfield images

Figure 2

The framework of the proposed automated cell line authentication system. In the data preparation stage, cell images are collected using the high-throughput IncuCyte microscopy technique from 30 cell lines and each cell image has two separate labels, e.g. cell line name and incubation time. Then, the deep learning network CLCNet learns the image-level features from the input cell images with their cell line labels and outputs predicted classes for test cell images. Once CLCNet is trained, the convolutional features of the training data are extracted to train CLRNet. CLRNet predicts the times of how long cell lines have been incubated simultaneously.

Back to article page